CN106599110A - Artificial intelligence-based voice search method and device - Google Patents
Artificial intelligence-based voice search method and device Download PDFInfo
- Publication number
- CN106599110A CN106599110A CN201611079405.4A CN201611079405A CN106599110A CN 106599110 A CN106599110 A CN 106599110A CN 201611079405 A CN201611079405 A CN 201611079405A CN 106599110 A CN106599110 A CN 106599110A
- Authority
- CN
- China
- Prior art keywords
- voice
- child
- search results
- artificial intelligence
- training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 45
- 238000012549 training Methods 0.000 claims description 39
- 238000012216 screening Methods 0.000 claims description 33
- 238000009432 framing Methods 0.000 claims description 22
- 230000000694 effects Effects 0.000 claims description 20
- 238000001514 detection method Methods 0.000 claims description 18
- 238000000605 extraction Methods 0.000 claims description 16
- 239000000284 extract Substances 0.000 claims description 3
- 238000001914 filtration Methods 0.000 abstract description 2
- 230000009323 psychological health Effects 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 238000003860 storage Methods 0.000 description 8
- 230000004069 differentiation Effects 0.000 description 7
- 230000004630 mental health Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000002411 adverse Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 210000003733 optic disk Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000009333 weeding Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/26—Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/43—Querying
- G06F16/432—Query formulation
- G06F16/433—Query formulation using audio data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/43—Querying
- G06F16/435—Filtering based on additional data, e.g. user or group profiles
- G06F16/436—Filtering based on additional data, e.g. user or group profiles using biological or physiological data of a human being, e.g. blood pressure, facial expression, gestures
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/54—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for retrieval
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/93—Discriminating between voiced and unvoiced parts of speech signals
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/78—Detection of presence or absence of voice signals
- G10L2025/783—Detection of presence or absence of voice signals based on threshold decision
- G10L2025/786—Adaptive threshold
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
- G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Acoustics & Sound (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Physiology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Provided are an artificial intelligence-based voice search method and device. The method comprises: identifying a voice input by a user to judge whether the voice is a child's voice; if a judgment result is yes, filtering search results obtained according to the voice; and feeding the filtered search results back to the user. In the present invention, a voice input by a user is identified to determine whether the user is a child user, and after the user is determined as the child user, search results are filtered, and search results about the child user can be differentially set, so sensitive contents which are included in the search results and are not suitable to be read by a child can be eliminated, and contents pushed to the child user contribute to the physical and psychological health of the child.
Description
Technical field
The present invention relates to technical field of information processing, more particularly to a kind of voice search method and dress based on artificial intelligence
Put.
Background technology
Artificial intelligence (Artificial Intelligence), english abbreviation is AI.It is study, be developed for simulation,
Extend and extend a new science of technology of theory, method, technology and the application system of the intelligence of people.Artificial intelligence is to calculate
One branch of machine science, it attempts the essence for understanding intelligence, and produce it is a kind of it is new can be in the way of human intelligence be similar
The intelligent machine made a response, the research in the field include robot, speech recognition, image recognition, natural language processing and specially
Family's system etc..Wherein, the most important aspect of artificial intelligence is exactly speech recognition technology.
With the continuous development of speech recognition technology, the range of application of speech recognition technology is also more and more extensive.With language
The lifting of sound recognition accuracy, increasing user are got used to using phonetic search mode, so as to avoid carrying out input through keyboard,
Simplify search procedure.At present, developing rapidly with intelligent terminal's technology, carries out the youngster of phonetic search using intelligent terminal
It is more and more, it should be noted that child is particularly preschooler and is likeed best by way of voice when using intelligent terminal
Interact.
During phonetic search is carried out, by speech recognition technology, the voice that child is input into is converted into into text, from
And the text for being converted is scanned for as search word.But the Search Results obtained based on search word usually contain it is irregular not
Neat information, as child user is mostly scanned for using the intelligent terminal of the head of a family, and cannot determine that whether the user is
Virgin user, leads to not carry out differentiation setting for the Search Results of child user colony.
The content of the invention
It is contemplated that at least solving one of technical problem in correlation technique to a certain extent.
For this purpose, first purpose of the present invention is to propose a kind of voice search method based on artificial intelligence, to realize
It is whether the identification of child user to user, for solving in prior art as user cannot be determined for child, leads to not
For its differentiation, the problem of Search Results is set.
Second object of the present invention is to propose a kind of voice searching device based on artificial intelligence.
Third object of the present invention is to propose another kind of voice searching device based on artificial intelligence.
Fourth object of the present invention is to propose a kind of non-transitorycomputer readable storage medium.
5th purpose of the present invention is to propose a kind of computer program.
It is that, up to above-mentioned purpose, first aspect present invention embodiment proposes a kind of phonetic search side based on artificial intelligence
Method, including：
The voice being input into user is identified, and judges whether the voice is child's voice；
If it is judged that being yes, the Search Results to being obtained according to the voice are screened；
Search Results after screening are fed back to into the user.
The voice search method based on artificial intelligence of the embodiment of the present invention, is known by the voice being input into user
Not, whether to determine the user as child user, after the user is determined for child user, then Search Results are sieved
Choosing, realizing the Search Results to child user carries out differentiation setting, does not meet child such that it is able to weed out Search Results
The sensitive content of reading so that the content pushed to child user has the physical and mental health using child.
It is that, up to above-mentioned purpose, second aspect present invention embodiment proposes a kind of phonetic search based on artificial intelligence and fills
Put, including：
Judge module, the voice for being input into user are identified, and judge whether the voice is child's voice；
Screening module, for being yes in the judged result of the judge module, to the search knot obtained according to the voice
Fruit is screened；
Feedback module, for the Search Results after screening are fed back to the user.
The voice searching device based on artificial intelligence of the embodiment of the present invention, is known by the voice being input into user
Not, whether to determine the user as child user, after the user is determined for child user, then Search Results are sieved
Choosing, realizing the Search Results to child user carries out differentiation setting, does not meet child such that it is able to weed out Search Results
The sensitive content of reading so that the content pushed to child user has the physical and mental health using child.
It is that, up to above-mentioned purpose, third aspect present invention embodiment proposes another kind of phonetic search based on artificial intelligence and fills
Put, including：Processor；For storing the memorizer of the processor executable；Wherein, the processor is configured to：
The voice being input into user is identified, and judges whether the voice is child's voice；If it is judged that be yes, to according to described
The Search Results that voice is obtained are screened；Search Results after screening are fed back to into the user.
To achieve these goals, fourth aspect present invention embodiment proposes a kind of non-transitory computer-readable storage
Medium, when the instruction in the storage medium is performed by the processor of server end so that server end is able to carry out one
The voice search method based on artificial intelligence is planted, methods described includes：The voice being input into user is identified, and judges described
Whether voice is child's voice；If it is judged that being yes, the Search Results to being obtained according to the voice are screened；After screening
Search Results feed back to the user.
To achieve these goals, fifth aspect present invention embodiment proposes a kind of computer program, when described
When instruction processing unit in computer program is performed, a kind of voice search method based on artificial intelligence, the side are performed
Method includes：The voice being input into user is identified, and judges whether the voice is child's voice；If it is judged that be it is yes, it is right
Screened according to the Search Results that the voice is obtained；Search Results after screening are fed back to into the user.
The additional aspect of the present invention and advantage will be set forth in part in the description, and partly will become from the following description
Obtain substantially, or recognized by the practice of the present invention.
Description of the drawings
Of the invention above-mentioned and/or additional aspect and advantage will become from the following description of the accompanying drawings of embodiments
It is substantially and easy to understand, wherein：
A kind of schematic flow sheet of voice search method based on artificial intelligence that Fig. 1 is provided by the embodiment of the present invention；
Flow process of the another kind that Fig. 2 is provided by the embodiment of the present invention based on the voice search method of artificial intelligence is illustrated
Figure；
A kind of training flow process of grader that Fig. 3 is provided by the embodiment of the present invention；
A kind of structural representation of voice searching device based on artificial intelligence that Fig. 4 is provided by the embodiment of the present invention；
Fig. 5 is a kind of structural representation of screening module provided in an embodiment of the present invention；
Fig. 6 is a kind of structural representation of judge module provided in an embodiment of the present invention.
Specific embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from start to finish
Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to for explaining the present invention, and be not considered as limiting the invention.
Below with reference to the accompanying drawings the voice search method based on artificial intelligence and device of the embodiment of the present invention are described.
A kind of schematic flow sheet of voice search method based on artificial intelligence that Fig. 1 is provided by the embodiment of the present invention.
Should be comprised the following steps based on the voice search method of artificial intelligence：
S101, the voice being input into user are identified, and judge whether voice is child's voice.
In the present embodiment, speech recognition modeling is provided with advance in a search engine, can be right by speech recognition modeling
The voice is input into by user is identified, to determine the classification of the voice.The speech recognition modeling can be machine learning model,
After the training to great amount of samples data and study, the machine learning model can be classified to the classification of voice.
If it is judged that being yes, then S102 is performed, if it is judged that being no, then terminate flow process.
S102, the Search Results to being obtained according to voice are screened.
Specifically, voice is carried out changing the content of text that can get voice, then using content of text as search
Key word, scanned for based on the key word of the search, obtain Search Results corresponding with the content of text of the voice.
Due to the information on the Internet it is more various, according to may include in the result that the content of text of language is searched
Sensitive content, or the bad network information.Due to having identified that in S101 voice is child's voice, illustrate to be input into the user of voice
For child user, if Search Results are not carried out screening that Search Results are pushed to the child user directly, this is may result in
Child user receives many unsound network consultings, and these bad network consultings or information can be to the physical and mental healths of child
Adversely affect.
In the present embodiment, in order to preferably provide service or network environment well to child user, according to language
Content of text get Search Results after, need to screen Search Results, be not suitable for weeding out in Search Results
The sensitive content of children's reading.Specifically, Search Results can be analyzed, in judging the Search Results, whether includes discomfort
Close children's reading sensitive content, if it is determined that go out Search Results go out protection have the sensitive content for being not suitable for children's reading, so that it may
To go out to weed out these sensitive contents from the Search Results.
S103, the Search Results after screening are fed back to into user.
As the sensitive content for being not suitable for children's reading is eliminated from initial Search Results, complete to Search Results
Screening, the Search Results after screening than being well suited to children's reading, then feed back to use by the content relative healths in Search Results
Family.
The voice search method based on artificial intelligence that the present embodiment is provided, is known by the voice being input into user
Not, whether to determine the user as child user, after the user is determined for child user, then Search Results are sieved
Choosing, realizing the Search Results to child user carries out differentiation setting, does not meet child such that it is able to weed out Search Results
The sensitive content of reading so that the content pushed to child user has the physical and mental health using child.
A kind of schematic flow sheet of voice search method based on artificial intelligence that Fig. 2 is provided by the embodiment of the present invention.
Should be comprised the following steps based on the voice search method of artificial intelligence：
S201, voice is carried out voice activity detection remove voice in it is quiet, obtain tested speech.
As user is during speaking, may there is pause, in order to the searching resource for reducing occupancy can be gathered
To the voice that is input into of user carry out voice activity detection (Voice Activity Detection, abbreviation VAD), with from sound
The prolonged quiet phase is recognized and is eliminated in sound signal stream, to reach the purpose for saving bandwidth resources, and advantageously reduces use
The time delay end to end that family is felt.
Specifically, framing can be carried out according to default first step length to voice, for example, can be right with 10ms as a frame
Every frame of voice carries out voice activity detection, quiet in voice to remove, will remove it is quiet after voice as tested speech.
S202, the acoustic featuress for extracting tested speech.
Specifically, framing can be carried out according to default second step length to tested speech, for example, the second step-length can be with
25ms is a frame, then according to default 3rd step-length carries out extraction acoustic featuress per frame to tested speech, for example, the 3rd step-length
Can be 10ms.In the present embodiment, acoustic featuress can be wave filter group (Filter bank40, abbreviation Fbank40) feature or
Person's mel-frequency cepstrum coefficient (Mel-Frequency Cepstral Coefficients, abbreviation MFCC) feature.
S203, the acoustic featuress of tested speech are input in object classifiers it are identified, judges that whether voice be
Child's voice.
Further, will remove it is quiet after voice as tested speech, in order to identify whether tested speech is child's voice,
Need to extract the acoustic featuress of the tested speech, then the acoustic featuress of the tested speech are input to the mesh for having trained
In mark grader, whether object classifiers can be identified to acoustic featuress to tested speech, to determine the voice as child's voice.
Specifically, after tested speech being input to object classifiers per frame corresponding acoustic featuress, object classifiers can
To give a mark to each frame, then the comprehensive score per frame carries out final judgement, in the present embodiment, can obtaining every frame
Point sum up it is average, according to plus and it is average after meansigma methodss as basis for estimation, when meansigma methodss are more than or equal to default
After threshold value, then may determine that the voice is child's voice, if being less than the threshold value, may determine that the non-child's voice of voice.
In the present embodiment, S204 is then performed when judging that voice is child's voice, when it is non-child's voice to judge voice, terminate
Flow process.
In the present embodiment, before voice is tested, need to be trained grader, to get target classification
Device.Fig. 3 is the classifier training flow process that provides in the embodiment of the present invention.As shown in figure 3, the classifier training includes following step
Suddenly：
S301, obtain sample voice for being trained to default grader.
In the present embodiment, the sound of child can be sampled as sample voice, the sample collected using these
Voice is trained to default grader.Preferably, the voice for being labeled as child's voice in history by manual sort can be gathered
As sample voice.
S302, sample voice is carried out voice activity detection remove training data in it is quiet, obtain train voice.
As user is during speaking, may there is pause, in order to reduce the searching resource of occupancy, get
Sample voice is carried out after sample voice voice activity detection remove training data in it is quiet, obtain train voice.
Specifically, framing can be carried out according to default first step length to sample voice, for example, can be with 10ms as one
Frame, carries out voice activity detection to every frame of sample voice, quiet in sample voice to remove, will remove it is quiet after sample
Voice is used as training voice.
S303, the acoustic featuress for extracting training voice.
Specifically, can be to training voice to carry out framing according to default second step length, for example, can be with 25ms as one
Frame, then according to default 3rd step-length extracts acoustic featuress to the every frame for training voice, wherein, acoustic featuress can be filtering
Device group (Filter bank40, abbreviation Fbank40) feature or mel-frequency cepstrum coefficient (Mel-Frequency
Cepstral Coefficients, abbreviation MFCC) feature.
S304, the acoustic featuress of training voice are input in grader and are trained, obtain object classifiers.
In the present embodiment, time recurrent neural network (Long Short-Time Memory can be based on
Projection, abbreviation LSTMP), deep neural network (Deep Neural Networks, abbreviation DNN) or circulation nerve
Network (Recurrent Neural Networks, abbreviation RNN) is modeled to grader.Calculation based on machine learning herein
Method builds grader and is not defined.
After the acoustic featuress for getting training voice, acoustic featuress are input in grader and are trained, work as classification
After the result stable convergence of device output, the training of the grader completes to obtain object classifiers.
S204, convert speech into content of text.
S205, scanned in the data base for setting up for child in advance according to content of text.
In order to ensure to provide preferably service or network environment for child, in the present embodiment, can build for child in advance
Stand some data bases, such as test item bank, nursery rhymes animation library, developmental game storehouse etc.
After content of text is converted speech into, can be entered in the data base for setting up for child in advance according to content of text
Line search, as these data bases are set up for children population, enables to the search that Search Results more meet child
In demand, and data base, information has been carried out screening relative healths.
If the content of correlation is not searched in the data base for setting up for child in advance, enter in other data bases
Line search, that is, perform S206；If the content of correlation is searched in the data base for setting up for child in advance, S207 is performed.
If S206, the content that correlation is not searched in the data base for setting up for child in advance, exist according to content of text
Scan for obtaining Search Results in other data bases.
When the content of text according to voice does not search the content of correlation in the data base pre-build for child, it is
Meet the demand of search user, can be scanned in other data bases according to content of text, think user search to phase
The content of pass.
S207, Search Results are carried out screening removing be not suitable for the sensitive content of children's reading.
Due to the information on the Internet it is more various, according to may include in the result that the content of text of language is searched
Sensitive content, or the bad network information.Due to having identified that in S203 voice is child's voice, illustrate to be input into the user of voice
For child user, if Search Results are not carried out screening that Search Results are pushed to the child user directly, this is may result in
Child user receives many unsound network consultings, and these bad network consultings or information can be to the physical and mental healths of child
Adversely affect.
In the present embodiment, in order to preferably provide service or network environment well to child user, to Search Results
Screened, to weed out the sensitive content for being not suitable for children's reading in Search Results.Specifically, Search Results can be carried out
Whether analysis, include the sensitive content for being not suitable for children's reading in judging the Search Results, if it is determined that goes out Search Results and goes out to protect
Shield has the sensitive content for being not suitable for children's reading, it is possible to go out to weed out these sensitive contents from the Search Results.
In the present embodiment, in order to preferably provide the network information, the knot searched in the data base for pre-building to child
Fruit can also be screened, and further increasing the safety of information.
S208, the Search Results after screening are fed back to into user.
As the sensitive content for being not suitable for children's reading is eliminated from initial Search Results, complete to Search Results
Screening, the Search Results after screening than being well suited to children's reading, then feed back to use by the content relative healths in Search Results
Family.
The voice search method based on artificial intelligence that the present embodiment is provided, is known by the voice being input into user
Not, whether to determine the user as child user, after the user is determined for child user, then Search Results are sieved
Choosing, realizing the Search Results to child user carries out differentiation setting, does not meet child such that it is able to weed out Search Results
The sensitive content of reading so that the content pushed to child user has the physical and mental health using child.
A kind of structural representation of voice searching device based on artificial intelligence that Fig. 4 is provided by the embodiment of the present invention.
Should be included based on the voice searching device of artificial intelligence：Judge module 11, screening module 12 and feedback module 13.
Wherein, judge module 11, the voice for being input into user are identified, and judge whether the voice is virgin
Sound.
Screening module 12, for being yes, to being obtained according to the voice search in the judged result of the judge module
As a result screened.
Feedback module 13, for the Search Results after screening are fed back to the user.
A kind of structural representation of screening module that Fig. 5 is provided for the present embodiment.As shown in figure 5, the screening module 12 is wrapped
Include：Converting unit 121, search unit 122 and screening unit 123.
Converting unit 121, for the voice is converted into content of text.
Search unit 122, for scanning for obtaining the Search Results according to the content of text.
Screening unit 123, removes the sensitive content for being not suitable for children's reading for carrying out screening to the Search Results.
Further, search unit 122, specifically for：
Scanned in the data base for setting up for child in advance according to the content of text.
If not searching the content of correlation in the database, according to the content of text in other data bases
Scan for, to obtain the Search Results.
A kind of structural representation of judge module that Fig. 6 is provided for the present embodiment.As shown in fig. 6, the judge module 11 is wrapped
Include：Acquiring unit 111, quiet removal unit 112, feature extraction unit 113, training unit 114 and identification judging unit 115.
Acquiring unit 111, for obtaining the sample voice for being trained to default grader.
Quiet removal unit 112, removes in the training data for voice activity detection is carried out to the sample voice
It is quiet, obtain train voice.
Feature extraction unit 113, for extracting the acoustic featuress of the training voice.
Training unit 114, is trained for the acoustic featuress of the training voice are input in the grader, obtains
To the object classifiers.
Further, quiet removal unit 112, specifically for：
According to default first step-length to the sample voice framing, and the every frame to the sample voice carries out voice work
Move quiet during detecting is removed per frame, obtain the training voice.
Feature extraction unit 113, specifically for：
Framing is carried out to the training voice according to default second step-length.
According to default 3rd step-length to framing after the training voice carry out acoustic featuress extraction.
Further, quiet removal unit 112, is additionally operable to carry out the voice voice activity detection removal voice
In it is quiet, obtain tested speech.
Feature extraction unit 113, is additionally operable to the acoustic featuress for extracting the tested speech.
Further, judge module 11, also include：
Identification judging unit 115, for being input in the object classifiers by the acoustic featuress of the tested speech
Row identification, to judge the voice whether as child's voice.
Further, quiet removal unit 112, also particularly useful for：
According to default first step-length to voice framing, and every frame is carried out quiet during voice activity detection is removed per frame
Sound, obtains the tested speech.
Feature extraction unit 113, also particularly useful for：
Framing is carried out to the tested speech according to default second step-length.
According to default 3rd step-length to framing after the tested speech carry out acoustic featuress extraction.
Identification judging unit 115, specifically for：
The tested speech is input in the object classifiers per the corresponding sound characteristic of frame and is given a mark.
Obtain the meansigma methodss of score of the tested speech per frame.
If the meansigma methodss exceed default threshold value, judge the tested speech as child's voice.
The voice searching device based on artificial intelligence that the present embodiment is provided, is known by the voice being input into user
Not, whether to determine the user as child user, after the user is determined for child user, then Search Results are sieved
Choosing, realizing the Search Results to child user carries out differentiation setting, does not meet child such that it is able to weed out Search Results
The sensitive content of reading so that the content pushed to child user has the physical and mental health using child.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
Example ", or the description of " some examples " etc. mean specific features with reference to the embodiment or example description, structure, material or spy
Point is contained at least one embodiment or example of the present invention.In this manual, to the schematic representation of above-mentioned term not
Identical embodiment or example must be directed to.And, the specific features of description, structure, material or feature can be with office
Combined in one or more embodiments or example in an appropriate manner.Additionally, in the case of not conflicting, the skill of this area
The feature of the different embodiments or example described in this specification and different embodiments or example can be tied by art personnel
Close and combine.
Additionally, term " first ", " second " are only used for describing purpose, and it is not intended that indicating or implying relative importance
Or the implicit quantity for indicating indicated technical characteristic.Thus, define " first ", the feature of " second " can express or
Implicitly include at least one this feature.In describing the invention, " multiple " are meant that at least two, such as two, three
It is individual etc., unless otherwise expressly limited specifically.
In flow chart or here any process described otherwise above or method description are construed as, expression includes
It is one or more for realizing custom logic function or process the step of the module of code of executable instruction, fragment or portion
Point, and the scope of the preferred embodiment of the present invention includes other realization, wherein the suitable of shown or discussion can not be pressed
Sequence, including according to involved function by it is basic simultaneously in the way of or in the opposite order, carry out perform function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
Expression or here logic described otherwise above and/or step, for example, are considered use in flow charts
In the order list of the executable instruction for realizing logic function, may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (as computer based system, the system including processor or other can hold from instruction
The system of row system, device or equipment instruction fetch execute instruction) use, or with reference to these instruction execution systems, device or set
It is standby and use.For the purpose of this specification, " computer-readable medium " can any can be included, store, communicate, propagate or pass
The dress that defeated program is used for instruction execution system, device or equipment or with reference to these instruction execution systems, device or equipment
Put.The more specifically example (non-exhaustive list) of computer-readable medium is including following：With the electricity that one or more connect up
Connecting portion (electronic installation), portable computer diskette box (magnetic device), random access memory (RAM), read only memory
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device, and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium can even is that the paper that can print described program thereon or other are suitable
Medium, because for example by carrying out optical scanning to paper or other media edlin, interpretation can then be entered or if necessary with which
His suitable method is processed to electronically obtain described program, is then stored in computer storage.
It should be appreciated that each several part of the present invention can be realized with hardware, software, firmware or combinations thereof.Above-mentioned
In embodiment, the software that multiple steps or method can be performed in memory and by suitable instruction execution system with storage
Or firmware is realizing.Such as, if realized with hardware with another embodiment, can be with following skill well known in the art
Any one of art or their combination are realizing：With for data signal is realized the logic gates of logic function from
Scattered logic circuit, the special IC with suitable combinational logic gate circuit, programmable gate array (PGA), scene can be compiled
Journey gate array (FPGA) etc..
Those skilled in the art are appreciated that to realize all or part of step that above-described embodiment method is carried
Suddenly the hardware that can be by program to instruct correlation is completed, and described program can be stored in a kind of computer-readable storage medium
In matter, the program upon execution, including one or a combination set of the step of embodiment of the method.
Additionally, each functional unit in each embodiment of the invention can be integrated in a processing module, it is also possible to
It is that unit is individually physically present, it is also possible to which two or more units are integrated in a module.Above-mentioned integrated mould
Block both can be realized in the form of hardware, it would however also be possible to employ the form of software function module is realized.The integrated module is such as
Fruit using in the form of software function module realize and as independent production marketing or use when, it is also possible to be stored in a computer
In read/write memory medium.
Storage medium mentioned above can be read only memory, disk or CD etc..Although having shown that above and retouching
Embodiments of the invention are stated, it is to be understood that above-described embodiment is exemplary, it is impossible to be interpreted as the limit to the present invention
System, one of ordinary skill in the art can be changed to above-described embodiment, change, replace and become within the scope of the invention
Type.
Claims (14)
1. a kind of voice search method based on artificial intelligence, it is characterised in that include：
The voice being input into user is identified, and judges whether the voice is child's voice；
If it is judged that being yes, the Search Results to being obtained according to the voice are screened；
Search Results after screening are fed back to into the user.
2. the voice search method based on artificial intelligence according to claim 1, it is characterised in that described to according to described
The Search Results that voice is obtained are screened, including：
The voice is converted into into content of text；
Scan for obtaining the Search Results according to the content of text；
The Search Results are carried out screening and removes the sensitive content for being not suitable for children's reading.
3. the voice search method based on artificial intelligence according to claim 2, it is characterised in that described according to the text
This content scans for obtaining the Search Results, including：
Scanned in the data base for setting up for child in advance according to the content of text；
If not searching the content of correlation in the database, carried out in other data bases according to the content of text
Search, to obtain the Search Results.
4. the voice search method based on artificial intelligence according to any one of claim 1-3, it is characterised in that described right
Before the voice is input into by user is identified, including：
Obtain the sample voice for being trained to default grader；
Voice activity detection is carried out to the sample voice and removes quiet in the training data, obtain training voice；
Extract the acoustic featuress of the training voice；
The acoustic featuress of the training voice are input in the grader and are trained, obtain the object classifiers.
5. the voice search method based on artificial intelligence according to claim 4, it is characterised in that described to the sample
Voice carries out voice activity detection and removes quiet in the training data, obtains training voice, including：
According to default first step-length to the sample voice framing, and speech activity is carried out to every frame of the sample voice detect
Survey quiet in removing per frame, obtain the training voice；
The acoustic featuress for extracting the training voice, including：
Framing is carried out to the training voice according to default second step-length,
According to default 3rd step-length to framing after the training voice carry out acoustic featuress extraction.
6. the voice search method based on artificial intelligence according to claim 4, it is characterised in that described defeated to user institute
The voice for entering is identified, and judges whether the voice is child's voice, including：
Voice activity detection is carried out to the voice and removes quiet in the voice, obtain tested speech；
Extract the acoustic featuress of the tested speech；
The acoustic featuress of the tested speech are input in the object classifiers and are identified, whether to judge the voice
For child's voice.
7. the voice search method based on artificial intelligence according to claim 6, it is characterised in that described to the voice
Carry out voice activity detection and remove quiet in the voice, obtain tested speech, including：
It is according to default first step-length to voice framing and quiet during voice activity detection removal is carried out to every frame per frame, obtain
To the tested speech；
The acoustic featuress for extracting the tested speech, including：
Framing is carried out to the tested speech according to default second step-length,
According to default 3rd step-length to framing after the tested speech carry out acoustic featuress extraction；
The acoustic featuress by the tested speech are identified in being input to the object classifiers, to judge the voice
Whether it is child's voice, including：
The tested speech is input in the object classifiers per the corresponding sound characteristic of frame and is given a mark；
Obtain the meansigma methodss of score of the tested speech per frame；
If the meansigma methodss exceed default threshold value, judge the tested speech as child's voice.
8. a kind of voice searching device based on artificial intelligence, it is characterised in that include：
Judge module, the voice for being input into user are identified, and judge whether the voice is child's voice；
Screening module, for being yes in the judged result of the judge module, the Search Results to being obtained according to the voice enter
Row screening；
Feedback module, for the Search Results after screening are fed back to the user.
9. the voice searching device based on artificial intelligence according to claim 8, it is characterised in that the screening module,
Including：
Converting unit, for the voice is converted into content of text；
Search unit, for scanning for obtaining the Search Results according to the content of text；
Screening unit, removes the sensitive content for being not suitable for children's reading for carrying out screening to the Search Results.
10. the voice searching device based on artificial intelligence according to claim 9, it is characterised in that the search unit,
Specifically for：
Scanned in the data base for setting up for child in advance according to the content of text；
If not searching the content of correlation in the database, carried out in other data bases according to the content of text
Search, to obtain the Search Results.
11. voice searching devices based on artificial intelligence according to any one of claim 8-10, it is characterised in that described
Judge module, including：
Acquiring unit, for obtaining the sample voice for being trained to default grader；
Quiet removal unit, it is quiet in the voice activity detection removal training data for carrying out to the sample voice,
Obtain training voice；
Feature extraction unit, for extracting the acoustic featuress of the training voice；
Training unit, is trained for the acoustic featuress of the training voice are input in the grader, obtains described
Object classifiers.
12. voice searching devices based on artificial intelligence according to claim 11, it is characterised in that the quiet removal
Unit, specifically for：
According to default first step-length to the sample voice framing, and speech activity is carried out to every frame of the sample voice detect
Survey quiet in removing per frame, obtain the training voice；
The feature extraction unit, specifically for：
Framing is carried out to the training voice according to default second step-length；
According to default 3rd step-length to framing after the training voice carry out acoustic featuress extraction.
13. voice searching devices based on artificial intelligence according to claim 11, it is characterised in that
The quiet removal unit, is additionally operable to carry out the voice voice activity detection and removes quiet in the voice, obtain
To tested speech；
The feature extraction unit, is additionally operable to the acoustic featuress for extracting the tested speech；
The judge module, also includes：
Identification judging unit, is identified for the acoustic featuress of the tested speech are input in the object classifiers,
To judge the voice whether as child's voice.
14. voice searching devices based on artificial intelligence according to claim 13, it is characterised in that the quiet removal
Unit, also particularly useful for：
It is according to default first step-length to voice framing and quiet during voice activity detection removal is carried out to every frame per frame, obtain
To the tested speech；
The feature extraction unit, also particularly useful for：
Framing is carried out to the tested speech according to default second step-length；
According to default 3rd step-length to framing after the tested speech carry out acoustic featuress extraction；
The identification judging unit, specifically for：
The tested speech is input in the object classifiers per the corresponding sound characteristic of frame and is given a mark；
Obtain the meansigma methodss of score of the tested speech per frame；
If the meansigma methodss exceed default threshold value, judge the tested speech as child's voice.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611079405.4A CN106599110A (en) | 2016-11-29 | 2016-11-29 | Artificial intelligence-based voice search method and device |
| US15/823,663 US10157619B2 (en) | 2016-11-29 | 2017-11-28 | Method and device for searching according to speech based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611079405.4A CN106599110A (en) | 2016-11-29 | 2016-11-29 | Artificial intelligence-based voice search method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106599110A true CN106599110A (en) | 2017-04-26 |
Family
ID=58595724
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611079405.4A Pending CN106599110A (en) | 2016-11-29 | 2016-11-29 | Artificial intelligence-based voice search method and device |
Country Status (2)
| Country | Link |
|---|---|
| US (1) | US10157619B2 (en) |
| CN (1) | CN106599110A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107610706A (en) * | 2017-09-13 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | The processing method and processing unit of phonetic search result |
| CN108766431A (en) * | 2018-06-06 | 2018-11-06 | 广东小天才科技有限公司 | Automatic awakening method based on voice recognition and electronic equipment |
| CN109087633A (en) * | 2018-08-23 | 2018-12-25 | 北京猎户星空科技有限公司 | Voice assessment method, device and electronic equipment |
| CN110660393A (en) * | 2019-10-31 | 2020-01-07 | 广东美的制冷设备有限公司 | Voice interaction method, device, equipment and storage medium |
| CN112231440A (en) * | 2020-10-09 | 2021-01-15 | 安徽讯呼信息科技有限公司 | Voice search method based on artificial intelligence |
| CN113486208A (en) * | 2021-06-09 | 2021-10-08 | 安徽沐峰数据科技有限公司 | Voice search equipment based on artificial intelligence and search method thereof |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107221326B (en) * | 2017-05-16 | 2021-05-28 | 百度在线网络技术（北京）有限公司 | Voice awakening method and device based on artificial intelligence and computer equipment |
| CN111916065B (en) * | 2020-08-05 | 2024-07-02 | 北京百度网讯科技有限公司 | Method and device for processing voice |
| CN112614514B (en) * | 2020-12-15 | 2024-02-13 | 中国科学技术大学 | Effective voice fragment detection method, related equipment and readable storage medium |
| CN113514544A (en) * | 2020-12-29 | 2021-10-19 | 大连理工大学 | Mobile robot pavement material identification method based on sound characteristics |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8417530B1 (en) * | 2010-08-20 | 2013-04-09 | Google Inc. | Accent-influenced search results |
| CN103151039A (en) * | 2013-02-07 | 2013-06-12 | 中国科学院自动化研究所 | Speaker age identification method based on SVM (Support Vector Machine) |
| CN104795067A (en) * | 2014-01-20 | 2015-07-22 | 华为技术有限公司 | Voice interaction method and device |
| CN105095406A (en) * | 2015-07-09 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Method and apparatus for voice search based on user feature |
| CN105677787A (en) * | 2015-12-29 | 2016-06-15 | 广州神马移动信息科技有限公司 | Information searching device and information searching method |
| US20160171109A1 (en) * | 2014-12-12 | 2016-06-16 | Ebay Inc. | Web content filtering |
Family Cites Families (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0127718B1 (en) * | 1983-06-07 | 1987-03-18 | International Business Machines Corporation | Process for activity detection in a voice transmission system |
| TW356548B (en) * | 1996-07-11 | 1999-04-21 | Sega Entpr Corp | Sound identifying device method of sound identification and the game machine using the said device |
| US6286001B1 (en) * | 1999-02-24 | 2001-09-04 | Doodlebug Online, Inc. | System and method for authorizing access to data on content servers in a distributed network |
| US6336117B1 (en) * | 1999-04-30 | 2002-01-01 | International Business Machines Corporation | Content-indexing search system and method providing search results consistent with content filtering and blocking policies implemented in a blocking engine |
| AU2001234011A1 (en) * | 2000-01-28 | 2001-08-07 | Sagi Cooper | Apparatus and method for accessing multimedia content |
| US20040128282A1 (en) * | 2001-03-07 | 2004-07-01 | Paul Kleinberger | System and method for computer searching |
| EP1402409A2 (en) * | 2001-06-08 | 2004-03-31 | W.W. Grainger, Inc. | System and method for retrieving information from an electronic catalog |
| US6731239B2 (en) * | 2002-01-18 | 2004-05-04 | Ford Motor Company | System and method for retrieving information using position coordinates |
| US7216121B2 (en) * | 2002-12-31 | 2007-05-08 | International Business Machines Corporation | Search engine facility with automated knowledge retrieval, generation and maintenance |
| US7672931B2 (en) * | 2005-06-30 | 2010-03-02 | Microsoft Corporation | Searching for content using voice search queries |
| US20080183694A1 (en) * | 2007-01-31 | 2008-07-31 | Daniel Cane | Method and system presenting search results using relationship information |
| US7882102B2 (en) * | 2007-09-10 | 2011-02-01 | Mitac International Corporation | Nearest-neighbor geographic search |
| US8160877B1 (en) * | 2009-08-06 | 2012-04-17 | Narus, Inc. | Hierarchical real-time speaker recognition for biometric VoIP verification and targeting |
| US20110302633A1 (en) * | 2010-06-07 | 2011-12-08 | Windschauer Robert J | Safe Internet Browser |
| US8484219B2 (en) * | 2010-09-21 | 2013-07-09 | Sony Computer Entertainment America Llc | Developing a knowledge base associated with a user that facilitates evolution of an intelligent user interface |
| US8768782B1 (en) * | 2011-06-10 | 2014-07-01 | Linkedin Corporation | Optimized cloud computing fact checking |
| KR101971008B1 (en) * | 2012-06-29 | 2019-04-22 | 삼성전자주식회사 | Control method for terminal using context-aware and terminal thereof |
| US20150287410A1 (en) * | 2013-03-15 | 2015-10-08 | Google Inc. | Speech and semantic parsing for content selection |
| US9489411B2 (en) * | 2013-07-29 | 2016-11-08 | Sybase, Inc. | High performance index creation |
| US10083237B2 (en) * | 2015-08-31 | 2018-09-25 | Google Llc | Protecting users from inappropriate sensitive or offensive search results |
| US20170337610A1 (en) * | 2016-05-20 | 2017-11-23 | Monroney Labels, LLC | Motor Vehicle Data Retrieval, Processing and Presentation Process |
-
2016
- 2016-11-29 CN CN201611079405.4A patent/CN106599110A/en active Pending
-
2017
- 2017-11-28 US US15/823,663 patent/US10157619B2/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8417530B1 (en) * | 2010-08-20 | 2013-04-09 | Google Inc. | Accent-influenced search results |
| CN103151039A (en) * | 2013-02-07 | 2013-06-12 | 中国科学院自动化研究所 | Speaker age identification method based on SVM (Support Vector Machine) |
| CN104795067A (en) * | 2014-01-20 | 2015-07-22 | 华为技术有限公司 | Voice interaction method and device |
| US20160171109A1 (en) * | 2014-12-12 | 2016-06-16 | Ebay Inc. | Web content filtering |
| CN105095406A (en) * | 2015-07-09 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Method and apparatus for voice search based on user feature |
| CN105677787A (en) * | 2015-12-29 | 2016-06-15 | 广州神马移动信息科技有限公司 | Information searching device and information searching method |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107610706A (en) * | 2017-09-13 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | The processing method and processing unit of phonetic search result |
| CN108766431A (en) * | 2018-06-06 | 2018-11-06 | 广东小天才科技有限公司 | Automatic awakening method based on voice recognition and electronic equipment |
| CN109087633A (en) * | 2018-08-23 | 2018-12-25 | 北京猎户星空科技有限公司 | Voice assessment method, device and electronic equipment |
| CN110660393A (en) * | 2019-10-31 | 2020-01-07 | 广东美的制冷设备有限公司 | Voice interaction method, device, equipment and storage medium |
| CN110660393B (en) * | 2019-10-31 | 2021-12-03 | 广东美的制冷设备有限公司 | Voice interaction method, device, equipment and storage medium |
| CN112231440A (en) * | 2020-10-09 | 2021-01-15 | 安徽讯呼信息科技有限公司 | Voice search method based on artificial intelligence |
| CN113486208A (en) * | 2021-06-09 | 2021-10-08 | 安徽沐峰数据科技有限公司 | Voice search equipment based on artificial intelligence and search method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| US20180151183A1 (en) | 2018-05-31 |
| US10157619B2 (en) | 2018-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106599110A (en) | Artificial intelligence-based voice search method and device | |
| CN110728997B (en) | Multi-modal depression detection system based on context awareness | |
| KR102601848B1 (en) | Device and method of data recognition model construction, and data recognition devicce | |
| CN1157710C (en) | Speech datas extraction | |
| CN105893478B (en) | A kind of tag extraction method and apparatus | |
| CN108320734A (en) | Audio signal processing method and device, storage medium, electronic equipment | |
| CN110910283A (en) | Method, device, equipment and storage medium for generating legal document | |
| CN105095415B (en) | The determination method and apparatus of network mood | |
| CN103765506A (en) | Method for tone/intonation recognition using auditory attention cues | |
| CN105023573A (en) | Speech syllable/vowel/phone boundary detection using auditory attention cues | |
| CN103744953A (en) | Network hotspot mining method based on Chinese text emotion recognition | |
| CN108899033B (en) | Method and device for determining speaker characteristics | |
| CN108345587A (en) | A kind of the authenticity detection method and system of comment | |
| Chittaragi et al. | Automatic text-independent Kannada dialect identification system | |
| CN106649849A (en) | Text information base building method and device and searching method, device and system | |
| CN108897732A (en) | Statement type recognition methods and device, storage medium and electronic device | |
| KR20190083143A (en) | Sensory evaluation method and apparatus | |
| CN110119443A (en) | A kind of sentiment analysis method towards recommendation service | |
| CN112418172A (en) | Multimode information fusion emotion analysis method based on multimode information intelligent processing unit | |
| CN107948730A (en) | Method, apparatus, equipment and storage medium based on picture generation video | |
| CN105183808A (en) | Problem classification method and apparatus | |
| CN106844340A (en) | News in brief generation and display methods, apparatus and system based on artificial intelligence | |
| CN109241993B (en) | Evaluation object emotion classification method and device integrating user and overall evaluation information | |
| CN109478405A (en) | Information processing equipment, information processing method and program | |
| Kello et al. | Scale-free networks in phonological and orthographic wordform lexicons |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170426 |